During Monday the 13th of August 1979 a depression named low ‘Y’ intensified and tracked quickly east across the eastern Atlantic. At midnight on Tuesday the 14th the low was close to Valentia Island in SW Ireland, with a minimum central pressure of around 979 hPa, making it one of the most intense lows for any August. The chart above (fig 1) is plotted from archived SYNOP observations that I’ve collected and also includes a background pressure field from the NCEP reanalysis project to fill in any gaps. The storm caught out the competitors in the Fastnet race and out of the 303 starters, only 86 finished. The Wikipedia article states that there were 194 retirements and 24 abandonments, five of which were “lost believed sunk”, and 15 sailors were drowned.
- In the main shipping forecast issued at 1355 BST on the 13th the forecast for Fasnet read:
SOLE, LUNDY, FASTNET
Southwesterly 4 to 6, increasing 6 or 7 for a time, veering westerly later. Occasional rain or showers. Moderate, locally poor, becoming good later.
- A gale warning was issued at ???? and broadcast at 1605 BST:
SOLE, FASTNET, SHANNON
Southwesterly gale force 8 imminent.
- A second gale warning was issued at 1805 BST and broadcast 1830 and 1906 BST it read
FINISTERRE, SOLE, FASTNET
Southwesterly gale force 8, increasing severe gale force 9 imminent.
- A third gale warning was issued at 2246 BST and broadcast at 2300 BST and read:
Southwesterly severe gales force 9, increasing storm force 10, imminent.
I’ve extracted these gale warnings from the PDF of the 1979 report ‘Fastnet race inquiry’ by the Royal Yachting Association and the Royal Ocean Racing Club. The scanning of the original report to create a PDF was not perfect. Here are a few of the more pertinent sections from the report that concern the weather and various warnings that were issued on that day.
2.16 Few competitors were listening to Radio 4 at 1600 on 13th when the first gale warning was broadcast. The 1355 shipping forecast had given little indication that there was a likelihood of gales and only 8% reported becoming aware of the severity of the storm between 1401 and 1600. It is not the general practice for those at sea to keep a continuous listening watch on Radio 4 so the value of gale warnings is limited. Although offshore racing yachts can be sailed through gales and it is generally accepted that winds of force 8 in open waters away from areas of fast tidal streams cause some discomfort but no real danger. Gale warnings are important to allow adequate precautions to be taken. The gale warning broadcast at 1830 and 1906 was the first to indicate that anything more than force 8 could be expected. It is unfortunate that the Meteorological Office issued the first two gale warnings just too late for inclusion in the shipping forecast. The force 10 warning broadcast at 2300 was the first to indicate the true nature of the winds which would be generated in area Fastnet by low Y
2.25 A detailed report on the weather situation and forecasting of the Fastnet storm has been produced by the Meteorological Office. The Inquiry has considered that report and much of the factual Information on the weather In this section is derived from it. Two articles by leading meteorologists on the Fastnet Race weather have also been published by the specialist press. These articles are believed to give excellent summaries of the weather for the period and they are therefore included at Annex 2B.
2.26 The Meteorological Office report makes it clear that as low Y crossed the North Atlantic on the 11th and 12th of August, the medium range forecasters were aware that it might deepen as it approached the British Isles and generate gale force winds. However there were very few ships in the Eastern Atlantic and South West Approaches to the British Isles sending weather reports when the low pressure area approached. On the 12th of August the Central Forecast Office received no indication that low Y was in fact deepening. On the evening of the 12th of August and the morning of the 13th of August there was no indication that strong or gale force winds would affect sea area Fastnet within the 24 hour forecast period. At this time low Y was deepening, but there were no reports available to indicate that it was doing so.
2.27 The Meteorological Office assessment of the actual weather on 14 August is based on very sparse information from the area in which the Fastnet Race fleet was sailing. The anemometer at the Kinsale platform which might have given a better record of wind strength was unserviceable at the time and there were very few ships in the area. Anemometers and barographs fitted to yachts are not calibrated and checked to the standards applied to official meteorological recording stations. It is hoped however, that with the benefit of all the data now available from yachts a further study which is being carried out by the Meteorological Office to improve numerical forecasting in the type of situation which prevailed will be fruitful.
- The report did included two articles about the Fastnet storm it its annex, one of them ‘The Storm’ was written by Alan Watts for the ‘Yachting World’ magazine of October 1979. He concludes:
Thus we begin to understand the Fastnet storm; a storm where the seaway was the governing factor in an extreme situation. In the Channel storm of 1956 where the winds grew along the Channel to the same ferocity as this year, there was not the same cross-sea problem as here. Yachts at sea were able to run under bare poles towing warps before the simple seaway, high as it was. This time the boats did not have a chance. No amount of seamanship would have prevented many of those which rolled, or were knocked down repeatedly, from succumbing to their fate. The cruel sea saw to that.
- The other article included in the annex ‘Tracking a killer storm’ was written by Robert B. Rice a meteorologist for the Sail magazine, who just writes a workmanlike story about how the storm developed and sticks just to the meteorology of the event, he finishes the article by saying:
This retreat (of low Y) from the scene allowed sea conditions to subside over the area, which permitted the widespread deployment of air sea rescue units to aid the stricken yachts. Had the storm lingered on for several days, the toll would very likely have been even more staggering.
Here are a few other articles that I’ve discovered about the Fastnet storm in the Met Mag and Weather Magazines. There seems to be a delayed reaction from both magazines concerning the storm. The first article wasn’t published till 18 months after the storm, it does seem strange that we can read an article about the ‘The violent thunderstorms of the 21/22 July 1907’, but when a severe gale drowns 15 sailors, with a further 10 related weather deaths on land, in a gale with 75 mph gusts in the middle of August, there’s little mention. of it, perhaps the cogs are moving but very slowly.
- In the February 1981 edition of the Met Mag there was an article entitled ‘Revised analyses and their effect on the fine-mesh forecast for the Fastnet storm’, and guess what when then added some observational data that was missing at the time the model got it spot on. Believe it or not I made that last line up before I read the article, which actually was not the case at all. The article, which oddly starts with the summary, says:
A particular combined analysis and forecast computation of the Meteorological Office fine-mesh numerical weather prediction model has been studied to discover why the ‘Fastnet low’ was not well predicted. The forecast was rerun several times, each run starting with a slightly changed and ‘improved’ analysis. It was found that a depression was predicted when the analysis was changed, and that this forecast depression could be deepened by increasing the number of analysed levels that were ‘improved’. With four levels of the original analysis changed a significant depression was forecast. The position of this depression was, however, some 200 nautical miles in error.
I wonder what solution the latest Cray supercomputer that the Met Office have invested in would think up nearly 40 years on?
- In the same edition there was another very interesting article entitled ‘Estimates of surface gust speeds using radar observations of showers’. It measured the speed of showers by the weather radar, and relates the speed to the maximum gusts. The summary for this read:
The speed of travel of shallow showers during the 1979 Fastnet Yacht Race, and on other occasions of strong wind, has been found to give a good indication of the peak surface gusts at exposed coastal locations. An objective echo tracking procedure is capable of determining the speed of travel of the showers provided that the radar data are available with a resolution of at least 5 km and 5 min.
The report shows a diagram of gusts as high as 65 knots in the Celtic Sea on the 14th of August. I don’t know if this method is still in use, something similar is now used to measure upper winds from cloud in satellite imagery.
- I found another article in the Met Mag from October 1981 entitled ‘The Fastnet storm—a forecaster’s viewpoint’ by A. Woodroffe which was detailed and frank, I doubt if an article like this from a Met Office employee would be allowed to see the light of day in these litigious days, how times have changed. In the conclusion to that article Woodroffe wrote:
(a) Guidance for 48 and 72 hours ahead from the 10-level model gave firm indications that low Y was likely to deepen as it approached western Ireland, although the surface pressure forecasts seriously underestimated the amount of deepening by 20-25 mb.
(d) Fine-mesh forecasts for 24 and 36 hours ahead failed to give advance warning of the sudden deepening and exceptional vigour of the low. Forecasts of the depth of low Y at the peak of the storm were generally in error by 20 mb or so.
- That last article prompted this one the Weather magazine a few years later in June 1984 edition entitled ‘Negligence in Weather Forecasting’ from Dennis Millington a solicitor who said about it:
…the account which was given raises many significant questions, the answers to which might prove embarrassing in the hands of a competent lawyer
- Another excellent article about the storm did appear rather belatedly in the Weather magazine for August 1997 entitled “The Fastnet storm of 1979: A mesoscale surface jet” by D. E. Pedgley, it said in part in its conclusion:
During the night of 13/14 August 1979, yachts in the Fastnet Race headed unknowingly towards the most damaging kind of wind system that can affect the British Isles – a mesoscale surface jet of hurricane force. This jet was comma-shaped and had formed on the southern side of a depression deepening rapidly in association with a mid-tropospheric dry intrusion of subsided air. Such depressions are common over the North Atlantic, notably in winter, but their mesoscale low-level jets have not been well-studied because of a lack of detailed observations.
I believe that in these days of satellite communications, ocean-going yachts are all equipped to download the latest surface wind and sea conditions directly from satellites links to the internet. As well as advances in communications, NWP models have also increased in power immensely, and are much more accurate, so the chances of sailors being caught out in this years Fastnet race are much reduced than they were in 1979. I notice that with this year’s Fastnet race, which starts on Sunday the 6th of August from Cowes, you can even watch how events unfold online (fig 2).
Interestingly in the latest forecast chart for midnight on Monday from the Met Office (fig 3), there is a low on the chart, thankfully this year its relative shallow at 1006 hPa, and much further north close to the Outer Hebrides rather than southwest Ireland, and by the looks of that chart, it might be poor visibility that’s the problem, rather than a force 10 southwesterly gale, as it was back in 1979.
All quoted excerpts from the Weather Magazine and the Met Mag are courtesy of the Royal Meteorological Society and the Meteorological Office.